Prioritizing drug targets for Alzheimer’s Disease using a Mendelian Randomization framework integrating neuroimaging and genomics

Abstract

AbstractBackgroundAlzheimer’s Disease is a progressive neurodegenerative disorder that currently has no disease‐modifying treatment, partly due to inefficiencies in identifying and validating drug targets. Drugs with genetic support are more likely to succeed in pharmaceutical trials. Recent studies show that in vivo neuroimaging can capture changes in brain pathology and function before onset of cognitive symptoms, thereby serving as biomarkers for early diagnosis. Pharmaceutical intervention may be more effective if treatment is started earlier in the disease course. We aim to integrate imaging and genomic data to identify and prioritize gene targets for Alzheimer’s Disease treatment.MethodWe apply Mendelian Randomization (MR) to investigate the druggable genome: 3,000 genes encoding proteins that can be targeted with molecular compounds. Gene expression mimics low‐level exposure to compounds in individuals.We examine the causal effect of expression and protein quantitative trait loci (eQTL and pQTL) information, from public datasets like GTex, in relation to markers of Alzheimer’s Disease risk and progression using GWAS summary statistics. The analyses are performed and replicated using two independent, publicly available datasets from ADNI and UK Biobank, including matched cases and controls. Genomic data will be imputed with a standard reference genome and harmonized across studies.ResultWe will describe the dataset and analysis pipeline needed to run and validate an MR experiment that identifies gene targets for disease‐modifying treatment based on eQTL and pQTL data. We will report results of descriptive and inferential statistical analyses associating these genes with disease risk, onset, and progression using related outcome variables such as structural imaging biomarkers, clinical measures and cognitive test scores. We propose a small subset of these gene targets with the strongest MR evidence as optimal candidates for drug development research, including potential drug repurposing opportunities.ConclusionThis study will identify gene targets for prioritization or repurposing in Alzheimer’s Disease drug development and clinical trials. The proposed targets are supported by genetic evidence of implication in disease risk, onset, and progression via association with neuroimaging features and cognitive scores. These results could aid pharmaceutical trials in terms of time and financial efficiency as well as increase the potential for success.